Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    material and object descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ad
    0.76
    ut
    0.66
    u
    0.63
    c
    0.61
    and
    0.58
    in
    0.57
    it
    0.57
    an
    0.56
    on
    0.55
    em
    0.53
    POSITIVE LOGITS
    ：
    0.44
     рассказывает
    0.42
     robin
    0.41
    뉩
    0.41
     oatmeal
    0.40
     paintbrush
    0.40
     obnov
    0.40
     watermelon
    0.40
    োল
    0.39
     violin
    0.39
    Act Density 0.207%

    No Known Activations